Recombinant HPV16 E7 vaccine plus CD137 agonism clears tumors in mice

A therapeutic vaccine engineered from the HPV16 E7 oncoprotein, delivered as a single recombinant protein rather than a mixture of synthetic peptides, has shown strong antitumor activity in mice, and its effect was substantially amplified when paired with an antibody that stimulates the co-stimulatory receptor 4-1BB, also known as CD137. The work, with Shisong Jiang as lead contact, offers a preclinical template for therapeutic HPV vaccination that avoids delivering intact, oncogenically active viral protein.

The rationale for a new antigen format stems from a persistent gap in HPV medicine. Prophylactic vaccines targeting the L1 capsid protein generate neutralizing antibodies and prevent new infections, but they offer no therapeutic benefit to people who already carry established HPV-related lesions or cancers. HPV16 and HPV18 account for roughly 70 percent of cervical cancer cases, and persistent infection with high-risk HPV types remains the leading cause of cervical and other anogenital malignancies. No therapeutic vaccine has been approved for patients with premalignant lesions or HPV-positive tumors, although preclinical and clinical efforts targeting the viral oncogenic proteins E6 and E7 have shown promise, particularly those that harness cell-mediated immunity.

Existing therapeutic approaches each carry drawbacks. DNA vaccines encoding the oncoproteins can activate cell-mediated immunity but raise safety concerns because E6 and E7 drive tumorigenesis. Wild-type protein vaccines cross-present antigen inefficiently, and epitope-based vaccines are restricted to patients with matching HLA types. Overlapping synthetic peptide and long synthetic peptide vaccines have achieved promising immunogenicity and clinical activity while using non-oncogenic peptide fragments, but peptide pools present practical problems: individual peptides are processed unevenly by antigen-presenting cells, some enter cells and access MHC molecules poorly, and synthesizing and quality-controlling many separate peptides adds regulatory complexity.

The platform tested here, called a recombinant overlapping peptide or ROP vaccine, addresses those limitations by genetically concatenating overlapping peptide fragments spanning the full wild-type protein sequence into a single recombinant construct. For HPV16 E7, a 98-amino-acid oncoprotein, the team divided the sequence into 35-amino-acid fragments with 10-amino-acid overlaps and joined them with LRMK linker motifs. The 35-residue length was chosen because long peptides are efficiently internalized by professional antigen-presenting cells and require intracellular processing before presentation, while the 10-residue overlap exceeds the typical length of minimal CD8 T cell epitopes, ensuring that every potential epitope is retained in at least one fragment. The LRMK motif serves as a cleavage site for cathepsin S, a protease abundantly expressed in professional antigen-presenting cells, so cleavage within endosomal compartments releases fragments for MHC loading.

Safety is built into the design. Wild-type E7 contains conserved regions critical for oncogenic activity: the N-terminal CR1 and CR2 domains, including the LXCXE motif that binds the retinoblastoma tumor suppressor family, and the C-terminal CR3 zinc-binding motif that supports structural stability and dimerization. These activities depend on the tertiary structure and intracellular expression of the full-length protein. Because ROP-E7 is delivered as an exogenous recombinant protein that is rapidly proteolyzed into short linear epitopes lacking higher-order structure, the functional domains required for pRb binding and zinc-dependent dimerization cannot be reconstituted. In silico analysis using the Immune Epitope Database suggested the construct may generate a broader array of potential T cell epitopes than wild-type E7, although the authors note that computational prediction does not directly assess epitope usage in vivo.

Immunization experiments in C57BL/6 mice bore out the design intent. Three subcutaneous doses of MPLA-adjuvanted ROP-E7, given at 14-day intervals, induced significantly stronger cellular responses than the same regimen with wild-type E7, measured by interferon-gamma ELISPOT after ex vivo restimulation with purified WT-E7 protein. Flow cytometry showed higher proportions of activated CD4 and CD8 T cells, and MHC class I tetramer staining for the immunodominant H-2Db-restricted E7 epitope RAHYNIVTF revealed greater expansion of E7-specific CD8 T cells in the ROP-E7 group. In a separate cohort immunized at 7-day intervals, splenocytes from ROP-E7-vaccinated mice lysed HPV16 E7-expressing TC-1 tumor cells far more effectively than splenocytes from controls.

Therapeutic efficacy was tested in the TC-1 model, in which mice were inoculated subcutaneously with 2 x 10^5 TC-1 cells on day 0 and vaccination began on day 5, when tumors were generally not yet palpable or only minimally palpable. The authors emphasize that this was an early-treatment setting rather than a model of large established tumors. Both vaccines inhibited tumor growth and prolonged survival relative to PBS and MPLA controls, with control tumors reaching roughly 1,500 cubic millimeters by day 22, WT-E7-treated tumors averaging about 700 cubic millimeters, and ROP-E7-treated tumors numerically smaller at under 500 cubic millimeters. The differences between the two vaccine groups did not reach statistical significance, so superiority of ROP-E7 over WT-E7 could not be formally established on these endpoints, though the trend prompted deeper analysis of the tumor immune microenvironment.

That analysis combined bulk RNA sequencing with two immune deconvolution algorithms, mMCP-counter and Digital Cell Quantification, plus flow cytometry of tumor-infiltrating lymphocytes. Tumors from ROP-E7-treated mice carried stronger T cell and CD8 T cell transcriptional signatures and significantly more CD8 tetramer-positive T cells than tumors from WT-E7-vaccinated mice, while intratumoral regulatory T cells showed a non-significant trend toward reduction and splenic Treg frequencies were unchanged, indicating that immune modulation was localized to the tumor. DCQ analysis associated ROP-E7 with CD8 effector memory and memory signatures, whereas WT-E7 tumors showed relatively higher CD4 memory and Treg signatures, and PBS control tumors carried higher stromal signatures including endothelial cells and cancer-associated fibroblasts.

Combination testing produced the study’s most striking results. Adding anti-PD-1 concurrently with the first vaccination did not improve tumor control over ROP-E7 alone under the dosing and schedule used, which the authors attribute to possible differences in antibody dose and timing relative to vaccine priming, since PD-1 signaling contributes to early CD8 T cell activation and clonal expansion. In contrast, combining ROP-E7 with a relatively low dose of 4-1BB agonistic antibody, which had limited activity as monotherapy, led to complete tumor eradication in the majority of treated mice within the initial 23-day analysis window and a trend toward prolonged survival. Extended follow-up to day 98 revealed the divergence clearly: all 15 ROP-E7 monotherapy mice eventually relapsed, yielding zero complete responses, while 10 of 15 combination-treated mice achieved complete regression and remained tumor-free. Transcriptomics showed the combination increased tumor-associated high endothelial venule signatures, structures that facilitate lymphocyte entry into tumors, and reduced Treg and exhaustion signatures, with tumor-infiltrating CD8 T cells expressing lower levels of PD-1 and LAG-3 than in the monotherapy group.

Durability, however, proved incomplete. Four tumor-free combination-treated mice were rechallenged with TC-1 cells on day 98; by day 120 one remained tumor-free while three developed tumors, indicating that long-term immune memory was partially retained but not uniformly protective. A second course of the combination regimen given on day 123 cleared recurrent tumors in two of the three mice with established recurrences, and a single ROP-E7 booster restored strong ex vivo cytotoxic activity in splenocytes from tumor-rejecting mice, showing that functional memory persisted but required antigenic re-stimulation. The authors caution that all findings come from a murine model that does not reproduce the heterogeneity of human HPV-associated malignancies, that safety assessment of 4-1BB agonism was limited to the absence of significant body weight loss despite known dose-dependent hepatotoxicity concerns with systemic agonists, that epitope breadth was inferred computationally rather than mapped experimentally, and that the mechanistic contribution of the LRMK linkers remains to be directly demonstrated. Within those limits, the study supports ROP-based antigen engineering, and its pairing with co-stimulatory immunotherapy, as a candidate strategy for therapeutic HPV vaccination.

Subject of Research: A recombinant overlapping peptide vaccine against HPV16 E7 combined with CD137 agonist immunotherapy for treating HPV-driven tumors in mice

Article Title: Therapeutic HPV16 E7 vaccination with CD137 agonism induces tumor regression and durable immunity in mice

Article References: Ou, Y., Lu, W., Zhao, H., Demetrious, C., Zhang, Q., Zheng, Q., Jiang, Y., Kok-Ting Wan, P., O’Neill, E., Zhang, L., & Jiang, S. (2026). Therapeutic HPV16 E7 vaccination with CD137 agonism induces tumor regression and durable immunity in mice. iScience, 29(11), Article 118018. https://doi.org/10.1016/j.isci.2026.118018

Image Credits: AI Generated

DOI: 10.1016/j.isci.2026.118018

Keywords: HPV16, E7 oncoprotein, therapeutic vaccine, recombinant overlapping peptide, 4-1BB, CD137 agonism, TC-1 tumor model, CD8 T cells, tumor microenvironment, immune memory, cancer immunotherapy, iScience